Disaster emergency assessment method and device fusing surveying and mapping data intelligent analysis

Abstract

The invention discloses a disaster situation emergency assessment method and device fusing surveying and mapping data intelligent analysis, and the device comprises a disaster situation surveying and mapping device which comprises a lifting column, an unmanned plane, a synthetic aperture radar sensor, a millimeter wave radar, an optical camera, a meteorological sensor and a gas sensor. The bottom plate is connected with the unmanned aerial vehicle, the millimeter-wave radar and the optical camera are installed on the left side of the lower end of the bottom plate, and the synthetic aperture radar sensor is installed on the right side of the lower end of the bottom plate. The disaster situation emergency intelligent assessment system comprises a machine learning module, a data fusion module, a geographic information system, a disaster situation emergency assessment comprehensive system and a disaster situation data processing module, the advantage complementary characteristics of a multi-modal data source are fused, immediate response of disaster situation emergency assessment is achieved, and the timeliness and accuracy of disaster situation assessment are improved.

Description

Technical Field

[0001] The present invention relates to the field of disaster emergency assessment, and in particular to a disaster emergency assessment method and device integrating intelligent analysis of surveying and mapping data. Background Art

[0002] At present, with the global climate change and the accelerated urbanization process, various natural disasters occur frequently, causing huge losses to the society and economy. Traditional disaster emergency assessment work relies on manual on-site investigation and post-event report summary, which is inefficient and has limited accuracy. In recent years, the rapid development of geographic information system GIS and remote sensing RS technology can quickly obtain the changes in topography and landforms in the disaster-stricken areas at the beginning of the disaster, providing key support for subsequent rescue operations.

[0003] The existing disaster emergency assessment system mainly collects real-time image data of the disaster area through satellite image analysis and drone aerial photography, and makes preliminary judgments based on the monitoring data transmitted by the ground sensor network.

[0004] However, the existing technology is constrained by many factors, such as the difficulty in obtaining remote sensing data due to cloud cover, the inability to effectively collect visible light images at night, and the lack of effective intelligent analysis tools for a large amount of raw data, resulting in slow processing speed and low accuracy. These problems affect the timeliness and accuracy of disaster assessment. To this end, a disaster emergency assessment method and device that integrates intelligent analysis of surveying and mapping data are proposed. Summary of the invention

[0005] The present invention provides the following technical solutions: The disaster emergency assessment method integrating intelligent analysis of surveying and mapping data includes the following steps: Step 1: Obtain an optical image of the ground surface through the optical camera at the bottom of the drone's bottom plate to realize remote sensing monitoring data collection of the target area. Obtain a radar image of the target area through the millimeter-wave radar at the bottom of the drone's bottom plate. The millimeter-wave radar collects data in different weather conditions. Obtain a high-resolution radar image through the synthetic aperture radar sensor at the bottom of the drone's bottom plate. The synthetic aperture radar sensor transmits and receives microwave signals and synthesizes a large-aperture antenna during flight to improve resolution. Step 2: The data acquisition module in the disaster data processing module is responsible for collecting terrain data, building image data and environmental parameter data from remote sensing mapping data, meteorological sensors and gas sensors. The data preprocessing module performs preprocessing on different types of collected data. For image data, it performs cropping and normalization operations; for sensor data, it performs data cleaning and standardization operations for subsequent analysis. The feature extraction module extracts effective features from the preprocessed data in the machine learning model. The model training and evaluation module uses labeled sample data to train the machine learning model, and uses cross-validation and other methods to evaluate the performance of the model. The model is optimized and adjusted according to the evaluation results. The disaster assessment module calculates the building damage index and the road traffic obstruction coefficient and the prediction of the diffusion range of dangerous gases based on the output results of the model and the integrated auxiliary information and comprehensive assessment of the disaster. Step 3: The remote sensing data is transmitted to the disaster emergency intelligent assessment system. The machine learning module processes the remote sensing data. The decision tree algorithm module identifies and classifies the degree of damage to the building structures in the disaster area based on the existing labeled data. The support vector machine performs binary or multi-classification on the terrain changes and the degree of damage to the building structures. The convolutional neural network is used to fuse the remote sensing mapping data and then extract features to determine the degree of damage. The recurrent neural network is used to process the time series data, analyze the wind speed time series environmental parameters, and predict the diffusion trend of dangerous gases. Step 4: The weighted balance module in the data fusion module assigns different weights to the remote sensing mapping data and the monitoring data of the meteorological sensor and the gas sensor, and then performs weighted average fusion. The Kalman filter module is used for dynamic data fusion, which integrates the remote sensing mapping data and the monitoring data of the meteorological sensor and the gas sensor in real time to obtain terrain and building information, thereby improving the real-time and accuracy of disaster assessment. Step 5: The geographic information system stores, manages, analyzes and visualizes various geographic spatial data. In disaster assessment, the information such as topographic changes and damage to building structures in the disaster-stricken area is superimposed with the population density, infrastructure and other layers in the GIS to intuitively display the distribution and impact range of the disaster; Step 6. The comprehensive disaster emergency assessment system integrates the above modules, including data collection, preprocessing, feature extraction, model training and evaluation, and disaster assessment functions, to achieve full-process automation or semi-automatic operation from data to disaster assessment results. The system has a user interface to facilitate emergency commanders to view and use disaster assessment results.

[0006] A disaster emergency assessment device integrating intelligent analysis of surveying and mapping data, including a disaster surveying and mapping device and a disaster emergency intelligent assessment system: The disaster mapping device includes a lifting column, an unmanned aerial vehicle, a synthetic aperture radar sensor, a millimeter wave radar, an optical camera, a meteorological sensor and a gas sensor. A base plate is provided at the lower end of the unmanned aerial vehicle, and the base plate is connected to the unmanned aerial vehicle. The millimeter wave radar and the optical camera are installed on the left side of the lower end of the base plate, and the synthetic aperture radar sensor is installed on the right side of the lower end of the base plate. The optical camera is used to obtain optical images of the surface of the earth. Data can be collected through the flight platform of the unmanned aerial vehicle to realize remote sensing monitoring of the target area. The data obtained by the optical camera is transmitted to the ground control station through wireless communication technology for subsequent image processing and analysis. The millimeter wave radar is used to obtain radar images of the target area. The radar sensor can work under different weather conditions, provide high-resolution distance and speed information, and transmit radar data Fusion with other remote sensing data to improve the accuracy of target detection and identification. For example, combining radar data with optical images can provide a more comprehensive understanding of the characteristics of the target. Synthetic aperture radar sensors are used to obtain high-resolution radar images. Synthetic aperture radar sensors can synthesize large-aperture antennas during flight by transmitting and receiving microwave signals to improve the resolution of radars. Synthetic aperture radar sensor remote sensing data can be used for real-time monitoring of surface changes, geological disaster warnings and other applications. Synthetic aperture radar sensors carried by drones can respond quickly and obtain timely remote sensing data. By carrying synthetic aperture radar sensors, millimeter-wave radars and optical cameras on the outside of drones, they are made compatible and work together to achieve complementary advantages of multiple remote sensing technologies and improve the efficiency and accuracy of remote sensing monitoring. The disaster emergency intelligent assessment system includes a machine learning module, a data fusion module, a geographic information system, a disaster emergency assessment integrated system and a disaster data processing module. The geographic information system can store, manage, analyze and visualize various geographic spatial data (such as terrain data, infrastructure distribution maps, etc.). In the disaster assessment, information such as terrain changes and damage to building structures in the disaster-stricken area can be superimposed with population density, infrastructure and other layers in the geographic information system to intuitively display the distribution and impact range of the disaster; The comprehensive disaster emergency assessment system integrates the above-mentioned modules, including data collection, preprocessing, feature extraction, model training and evaluation, disaster assessment and other functions, to achieve full-process automation or semi-automatic operation from data to disaster assessment results. The system also has a user interface to facilitate emergency commanders to view and use disaster assessment results in order to make scientific decisions.

[0007] Preferably, the disaster data processing module includes a data acquisition module, a data preprocessing module, a feature extraction module, a model training and evaluation module and a disaster assessment module. The data acquisition module is responsible for collecting various data from equipment such as drone remote sensing data and ground sensors, including terrain data, building image data, environmental parameter data, etc. This module must ensure the integrity and accuracy of the data and be able to perform preliminary processing on data in different formats. The data preprocessing module preprocesses the different types of collected data, performs cropping, normalization and other operations on image data; performs data cleaning (removing outliers, etc.) and standardization and other operations on sensor data for subsequent analysis. The feature extraction module extracts effective features from the preprocessed data in the machine learning model. For example, edge features and texture features are extracted from building images, and features related to the disaster (such as the relationship between temperature and the diffusion rate of hazardous gases) are extracted from environmental parameter data. The model training and evaluation module uses labeled sample data to train the machine learning model, and uses cross-validation and other methods to evaluate the performance of the model. The model is optimized and adjusted according to the evaluation results to improve the accuracy of the model in identifying and classifying the characteristics of the disaster-stricken areas. The disaster assessment module comprehensively assesses the disaster based on the output results of the model (such as classification of the degree of damage to the building structure, terrain changes, etc.) and integrated auxiliary information (population density, infrastructure distribution map, etc.), including calculating the building damage index, road traffic obstruction coefficient, and prediction of the diffusion range of hazardous gases.

[0008] Preferably, the machine learning module includes a decision tree algorithm module, a support vector machine, a convolutional neural network and a recurrent neural network. The decision tree algorithm module is used to identify and classify the degree of damage to the building structures in the disaster area based on the existing labeled data (such as building samples with known degree of damage). The decision tree is gradually classified by judging the features (such as the appearance features of the building, the surrounding environment, etc.). The support vector machine has a good performance for the binary or multi-classification problems of terrain changes and the degree of damage to the building structure. It separates data of different categories by finding an optimal hyperplane. The convolutional neural network has great advantages in processing image data (such as satellite images, drone images). For example, the pre-trained CNN model (such as ResNet, VGG, etc.) can be used to extract features from building images to determine the degree of damage. The data fusion module is effective in processing time series data, such as analyzing time series environmental parameters such as wind speed, predicting the diffusion trend of hazardous gases, etc.

[0009] Preferably, the data fusion module includes a weighted balancing module and a Kalman filtering module. The weighted balancing module assigns different weights to data from different sources (such as satellite images and ground sensor data) according to their reliability, and then performs weighted average fusion. The Kalman filtering module is suitable for dynamic data fusion, such as real-time fusion of environmental parameters returned by the ground sensor network and terrain and building information obtained by remote sensing data, so as to improve the real-time and accuracy of disaster assessment.

[0010] Preferably, a driving motor is provided at the upper end of the lifting column, and the driving motor is connected to the lifting column, and a transparent top box is installed at the output end of the driving motor, and the transparent top box is connected to the output end of the driving motor. The transparent top box protects the UAV. At the same time, through its transparent characteristics, the optical camera of the UAV can penetrate the transparent top box to perform remote sensing monitoring of the surrounding area, and cooperate with the driving motor to control the rotation of the transparent top box to increase the range of remote sensing monitoring. The lifting column allows the height of the transparent top box and the UAV to be raised and lowered to meet usage requirements.

[0011] Preferably, a wireless charging receiving device is provided at the lower end of the base plate and is connected to the base plate. The wireless charging receiving device receives the energy signal sent from the wireless charging transmitter through wireless charging technologies such as electromagnetic induction and magnetic resonance, and converts it into electrical energy that can be used for charging.

[0012] Preferably, a wireless charger is provided inside the transparent top box below the wireless charging receiving device, and the wireless charger is connected to the transparent top box. The wireless charger is a device that uses wireless charging technologies such as electromagnetic induction and magnetic resonance to wirelessly transmit electrical energy to a device that supports wireless charging. The wireless charger serves as an energy transmitting end, converts electrical energy into magnetic field energy and transmits it outward; the wireless charging receiving device serves as an energy receiving end, which can receive magnetic field energy and convert it into electrical energy, thereby realizing wireless transmission and conversion of electrical energy, which is convenient for replenishing electrical energy for the drone.

[0013] Preferably, a fixing ring is provided on the outside of the lifting column, and the fixing ring is connected to the lifting column, and the meteorological sensor and the gas sensor are respectively arranged on both sides of the outside of the fixing ring, and the fixing ring is used to install the meteorological sensor and the gas sensor. The meteorological sensor can monitor a variety of meteorological parameters in real time, such as temperature, humidity, air pressure, wind speed, wind direction, precipitation, etc. These data are very important for understanding the meteorological conditions when the disaster occurs. For example, in flood disasters, precipitation data can help predict the scale and development trend of the flood; in forest fires, wind speed and wind direction data are of key significance for predicting the spread direction and speed of the fire. By performing intelligent analysis on the data collected by the meteorological sensor, a disaster warning model can be established to predict the occurrence and development of disasters in advance. In the process of disaster emergency response, the data of the meteorological sensor can provide support for decision-making, and the gas sensor can monitor Various harmful gases in the air, such as sulfur dioxide, nitrogen oxides, carbon monoxide, ozone, etc., may produce a large amount of harmful gases when disasters occur, such as industrial accidents and fires. Gas sensors can monitor the concentration of these gases in real time and provide data support for environmental monitoring and pollution assessment. When a fire occurs, gas sensors can detect fire characteristic gases such as smoke and carbon monoxide in the air, thereby achieving early warning of the fire. The harmful gas concentration data monitored by the gas sensor can be used to assess the risks to human health. Meteorological sensors and gas sensors play an important role in the disaster emergency assessment method of intelligent analysis of surveying and mapping data. They can provide key data and technical support for disaster warning, risk assessment, decision support, environmental monitoring, health protection, etc., which will help improve the efficiency and level of disaster emergency management and reduce disaster losses.

[0014] Preferably, a base is provided at the lower end of the lifting column, and the base is connected to the lifting column, and four mounting holes evenly distributed along a ring are provided inside the base, and the mounting holes are integrally formed with the base, the base is used for installing and fixing the lifting column, and the mounting holes are used for the fixing parts to pass through, so as to fix the base to the ground or other positions.

[0015] In summary, compared with the prior art, the present invention has the following beneficial effects: This invention adopts the method of combining UAV with synthetic aperture radar and ground sensor network equipment. The synthetic aperture radar can penetrate the clouds and obtain surface information in bad weather, ensuring that there is still a reliable source of remote sensing data when optical remote sensing is limited. The ground sensor network is not affected by cloud cover and can obtain environmental parameters such as temperature, humidity, wind speed and so on in real time in the disaster area, thus supplementing the deficiency of remote sensing data. Through the two data fusion modules of weighted averaging method and Kalman filtering module, data from different sources, including SAR data which is less affected by clouds and ground sensor data, are effectively fused to improve the integrity and availability of data, thus overcoming the logarithm of cloud cover and rain cover. The convolutional neural network in the machine learning algorithm processes various types of data, whether it is visible light images or other types of images that can be obtained at night, and extracts features for analysis to ensure effective feature recognition and disaster assessment of the disaster-stricken areas at night; the comprehensive disaster emergency assessment system integrates various modules to achieve full-process automation or semi-automatic operation from data collection to disaster assessment results, avoiding the problems of inefficiency and low accuracy when manually processing large amounts of raw data. At the same time, the geographic information system effectively manages and analyzes various data, improves the efficiency of data utilization, and helps to improve the timeliness and accuracy of disaster assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the steps of the disaster emergency assessment method integrating intelligent analysis of surveying and mapping data; Figure 2 It is a structural stereogram of a disaster surveying and mapping device that is a disaster emergency assessment device that integrates intelligent analysis of surveying and mapping data; Figure 3 It is a cross-sectional view of the transparent top box structure of a disaster surveying and mapping device of a disaster emergency assessment device integrating intelligent analysis of surveying and mapping data; Figure 4 It is a top view of the UAV structure of the disaster mapping device of the disaster emergency assessment device integrating the intelligent analysis of surveying and mapping data; Figure 5 It is a top view of the UAV structure of the disaster mapping device of the disaster emergency assessment device integrating the intelligent analysis of surveying and mapping data; Figure 6 It is a schematic diagram of a disaster emergency intelligent assessment system that integrates a disaster emergency assessment device with intelligent analysis of surveying and mapping data.

[0017] Description of reference numerals: 1. Lifting column; 2. Base; 3. Mounting hole; 4. Driving motor; 5. Transparent top box; 6. Fixed ring; 7. Meteorological sensor; 8. Gas sensor; 9. Drone; 10. Wireless charger; 11. Bottom plate; 12. Wireless charging receiving device; 13. Synthetic aperture radar sensor; 14. Millimeter wave radar; 15. Optical camera; 16. Machine learning module; 17. Decision tree algorithm module; 18. Support vector machine; 19. Convolutional neural network; 20. Recurrent neural network; 21. Data fusion module; 22. Weighted balance module; 23. Kalman filter module; 24. Geographic information system; 25. Comprehensive disaster emergency assessment system; 26. Data acquisition module; 27. Data preprocessing module; 28. Feature extraction module; 29. ​​Model training and evaluation module; 30. Disaster assessment module; 31. Disaster data processing module. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0019] See also Figure 1 The disaster emergency assessment method for integrating surveying and mapping data intelligent analysis provided by the present invention comprises the following steps: Step 1: The optical image of the ground surface is acquired through the optical camera 15 at the lower end of the bottom plate 11 of the drone 9 to realize remote sensing monitoring data collection of the target area. The radar image of the target area is acquired through the millimeter wave radar 14 at the lower end of the bottom plate 11 of the drone 9. The millimeter wave radar 14 collects data in different weather conditions. The high-resolution radar image is acquired through the synthetic aperture radar sensor 13 at the lower end of the bottom plate 11 of the drone 9. The synthetic aperture radar sensor 13 transmits and receives microwave signals and synthesizes a large aperture antenna during flight to improve the resolution. Step 2: The data acquisition module 26 in the disaster data processing module 31 is responsible for collecting terrain data, building image data and environmental parameter data from remote sensing mapping data, meteorological sensors 7 and gas sensors 8. The data preprocessing module 27 performs preprocessing on the different types of data collected. For image data, it performs cropping and normalization operations; for sensor data, it performs data cleaning and standardization operations for subsequent analysis. The feature extraction module 28 extracts effective features from the preprocessed data in the machine learning model. The model training and evaluation module 29 uses the labeled sample data to train the machine learning model, and uses cross-validation and other methods to evaluate the performance of the model. The model is optimized and adjusted according to the evaluation results. The disaster assessment module 30 calculates the building damage index and the road traffic obstruction coefficient and the prediction of the diffusion range of dangerous gases based on the output results of the model and the integrated auxiliary information and comprehensive assessment of the disaster. Step 3: The remote sensing data is transmitted to the disaster emergency intelligent assessment system. The machine learning module 16 processes the remote sensing data. The decision tree algorithm module 17 identifies and classifies the degree of damage to the building structures in the disaster area based on the existing labeled data. The support vector machine 18 performs binary or multi-classification of terrain changes and the degree of damage to the building structures. The convolutional neural network 19 is used to fuse the remote sensing mapping data and then extract features to determine the degree of damage. The recurrent neural network 20 is used to process the time series data, analyze the wind speed time series environmental parameters, and predict the diffusion trend of dangerous gases. Step 4: The weighted balance module 22 in the data fusion module 21 assigns different weights to the remote sensing mapping data and the monitoring data of the meteorological sensor 7 and the gas sensor 8, and then performs weighted average fusion. The Kalman filter module 23 is used for dynamic data fusion, which integrates the remote sensing mapping data and the monitoring data of the meteorological sensor 7 and the gas sensor 8 in real time to obtain terrain and building information, thereby improving the real-time and accuracy of disaster assessment. Step 5: The geographic information system 24 stores, manages, analyzes and visualizes various geographic spatial data. In disaster assessment, the information such as topographic changes and damage to building structures in the disaster-stricken area is superimposed with the population density, infrastructure and other layers in the GIS to intuitively display the distribution and impact range of the disaster; Step 6. The disaster emergency assessment integrated system 25 integrates the above modules, including data collection, preprocessing, feature extraction, model training and evaluation, and disaster assessment functions, to achieve full-process automation or semi-automatic operation from data to disaster assessment results. The system has a user interface to facilitate emergency command personnel to view and use disaster assessment results.

[0020] See also Figure 2 , Figure 3 , Figure 4 and Figure 5 , a disaster emergency assessment device integrating intelligent analysis of surveying and mapping data, including a disaster surveying and mapping device and a disaster emergency intelligent assessment system; The disaster mapping device includes a lifting column 1, an unmanned aerial vehicle 9, a synthetic aperture radar sensor 13, a millimeter wave radar 14, an optical camera 15, a meteorological sensor 7 and a gas sensor 8. A base plate 11 is provided at the lower end of the unmanned aerial vehicle 9, and the base plate 11 is connected to the unmanned aerial vehicle 9. The millimeter wave radar 14 and the optical camera 15 are installed on the left side of the lower end of the base plate 11, and the synthetic aperture radar sensor 13 is installed on the right side of the lower end of the base plate 11. The optical camera 15 is used to obtain optical images of the surface of the earth. Data can be collected through the flight platform of the unmanned aerial vehicle 9 to realize remote sensing monitoring of the target area. The data obtained by the optical camera 15 is transmitted to the ground control station through wireless communication technology for subsequent image processing and analysis. The millimeter wave radar 14 is used to obtain radar images of the target area. The radar sensor can work under different weather conditions and provide high-resolution distance and speed information. The radar data is integrated with other remote sensing data to improve the accuracy of target detection and identification. For example, the radar data is combined with the optical image to more fully understand the characteristics of the target. The synthetic aperture radar sensor 13 is used to obtain high-resolution radar images. The synthetic aperture radar sensor 13 can synthesize a large aperture antenna during flight by transmitting and receiving microwave signals to improve the resolution of the radar. The remote sensing data of the synthetic aperture radar sensor 13 can be used for real-time monitoring of surface changes, geological disaster warning and other applications. The synthetic aperture radar sensor 13 carried by the UAV 9 can respond quickly and obtain timely remote sensing data. By carrying the synthetic aperture radar sensor 13, millimeter wave radar 14 and optical camera 15 on the outside of the UAV 9, they are compatible and work together to achieve complementary advantages of multiple remote sensing technologies and improve the efficiency and accuracy of remote sensing monitoring. See also Figure 6 The disaster emergency intelligent assessment system includes a machine learning module 16, a data fusion module 21, a geographic information system 24, a disaster emergency assessment integrated system 25 and a disaster data processing module 31. The geographic information system 24 can store, manage, analyze and visualize various geographic spatial data (such as terrain data, infrastructure distribution maps, etc.). In the disaster assessment, the terrain changes in the disaster-stricken area, the damage to the building structure and other information can be superimposed with the population density, infrastructure and other layers in the geographic information system 24 to intuitively display the distribution and impact range of the disaster. The disaster emergency assessment integrated system 25 integrates the above-mentioned modules, including data collection, preprocessing, feature extraction, model training and evaluation, disaster assessment and other functions, to achieve full-process automation or semi-automatic operation from data to disaster assessment results. The system also has a user interface to facilitate emergency commanders to view and use disaster assessment results in order to make scientific decisions.

[0021] The disaster data processing module 31 includes a data acquisition module 26, a data preprocessing module 27, a feature extraction module 28, a model training and evaluation module 29 and a disaster assessment module 30. The data acquisition module 26 is responsible for collecting various data from remote sensing data of the drone 9 and ground sensors and other equipment, including terrain data, building image data, environmental parameter data, etc. This module must ensure the integrity and accuracy of the data and be able to perform preliminary processing on data in different formats. The data preprocessing module 27 preprocesses the different types of data collected, and performs operations such as cropping and normalization on image data; for sensor data, data cleaning (removing outliers, etc.) and standardization are performed to facilitate subsequent analysis. The feature extraction module 28 extracts effective features from the preprocessed data in the machine learning model. For example, edge features and texture features are extracted from building images, and features related to the disaster (such as the relationship between temperature and the diffusion rate of hazardous gases) are extracted from environmental parameter data. The model training and evaluation module 29 uses labeled sample data to train the machine learning model, and uses cross-validation and other methods to evaluate the performance of the model. The model is optimized and adjusted according to the evaluation results to improve the accuracy of the model in identifying and classifying the characteristics of the disaster-stricken areas. The disaster assessment module 30 comprehensively assesses the disaster based on the output results of the model (such as classification of the degree of damage to the building structure, terrain changes, etc.) and integrated auxiliary information (population density, infrastructure distribution map, etc.), including calculating the building damage index, road traffic obstruction coefficient, and prediction of the diffusion range of hazardous gases.

[0022] See also Figure 6 The machine learning module 16 includes a decision tree algorithm module 17, a support vector machine 18, a convolutional neural network 19 and a recurrent neural network 20. The decision tree algorithm module 17 is used to identify and classify the degree of damage to the building structure in the disaster area based on the existing labeled data (such as building samples with known damage degrees). The decision tree is gradually classified by judging the features (such as the appearance features of the building, the surrounding environment, etc.). The support vector machine 18 has a good performance for the binary or multi-classification problems of terrain changes and the degree of damage to the building structure. It separates data of different categories by finding an optimal hyperplane. The convolutional neural network 19 has great advantages in processing image data (such as satellite images, drone images). For example, pre-trained CNN models (such as ResNet, VGG, etc.) can be used to extract features from building images and then judge their degree of damage. The data fusion module 21 is effective in processing time series data, such as analyzing time series environmental parameters such as wind speed, predicting the diffusion trend of hazardous gases, etc.

[0023] See also Figure 6The data fusion module 21 includes a weighted balance module 22 and a Kalman filter module 23. The weighted balance module 22 assigns different weights to data from different sources (such as satellite images and ground sensor data) according to their reliability, and then performs weighted average fusion. The Kalman filter module 23 is suitable for dynamic data fusion, such as real-time fusion of environmental parameters returned by the ground sensor network and terrain and building information obtained by remote sensing data, so as to improve the real-time and accuracy of disaster assessment.

[0024] See also Figure 2 , Figure 3 and Figure 4 A driving motor 4 is provided at the upper end of the lifting column 1, and the driving motor 4 is connected to the lifting column 1. A transparent top box 5 is installed at the output end of the driving motor 4, and the transparent top box 5 is connected to the output end of the driving motor 4. The transparent top box 5 plays a protective role for the UAV 9. At the same time, through its transparent characteristics, the optical camera 15 of the UAV 9 can penetrate the transparent top box 5 to perform remote sensing monitoring of the surrounding area, and cooperate with the driving motor 4 to control the rotation of the transparent top box 5 to increase the range of remote sensing monitoring. The lifting column 1 allows the height of the transparent top box 5 and the UAV 9 to be adjusted to meet the use requirements.

[0025] See also Figure 4 and Figure 5 A wireless charging receiving device 12 is provided at the lower end of the base plate 11, and the wireless charging receiving device 12 is connected to the base plate 11. The wireless charging receiving device 12 receives the energy signal sent by the wireless charging transmitter through wireless charging technologies such as electromagnetic induction and magnetic resonance, and converts it into electrical energy that can be used for charging.

[0026] See also Figure 3 A wireless charger 10 is arranged inside the transparent top box 5 below the wireless charging receiving device 12, and the wireless charger 10 is connected to the transparent top box 5. The wireless charger 10 is a device that uses wireless charging technologies such as electromagnetic induction and magnetic resonance to wirelessly transmit electrical energy to a device that supports wireless charging. The wireless charger 10 serves as an energy transmitting end, converts electrical energy into magnetic field energy and transmits it outward; the wireless charging receiving device 12 serves as an energy receiving end, which can receive magnetic field energy and convert it into electrical energy, thereby realizing wireless transmission and conversion of electrical energy, thereby facilitating the replenishment of electrical energy for the drone 9.

[0027] See also Figure 2A fixing ring 6 is arranged on the outside of the lifting column 1, and the fixing ring 6 is connected to the lifting column 1. A meteorological sensor 7 and a gas sensor 8 are arranged on both sides of the outside of the fixing ring 6 respectively. The fixing ring 6 is used to install the meteorological sensor 7 and the gas sensor 8. The meteorological sensor 7 can monitor a variety of meteorological parameters in real time, such as temperature, humidity, air pressure, wind speed, wind direction, precipitation, etc. These data are very important for understanding the meteorological conditions when the disaster occurs. For example, in a flood disaster, precipitation data can help predict the scale and development trend of the flood; in a forest fire, wind speed and wind direction data are of key significance for predicting the spread direction and speed of the fire. By intelligently analyzing the data collected by the meteorological sensor 7, a disaster warning model can be established to predict the occurrence and development of the disaster in advance. In the process of disaster emergency response, the data of the meteorological sensor 7 can provide support for decision-making, and the gas sensor 8 can monitor Measure various harmful gases in the air, such as sulfur dioxide, nitrogen oxides, carbon monoxide, ozone, etc. When a disaster occurs, such as an industrial accident, a fire, etc., a large amount of harmful gases may be produced. The gas sensor 8 can monitor the concentration of these gases in real time and provide data support for environmental monitoring and pollution assessment. When a fire occurs, the gas sensor 8 can detect smoke, carbon monoxide and other fire characteristic gases in the air, thereby achieving early warning of the fire. The harmful gas concentration data monitored by the gas sensor 8 can be used to assess the risk to human health. Meteorological sensors 7 and gas sensors 8 play an important role in the disaster emergency assessment method of intelligent analysis of surveying and mapping data. They can provide key data and technical support for disaster warning, risk assessment, decision support, environmental monitoring, health protection, etc., which will help improve the efficiency and level of disaster emergency management and reduce disaster losses.

[0028] See also Figure 2 A base 2 is provided at the lower end of the lifting column 1, and the base 2 is connected to the lifting column 1. Four mounting holes 3 evenly distributed along a ring are provided inside the base 2, and the mounting holes 3 are integrally formed with the base 2. The base 2 is used for installing and fixing the lifting column 1, and the mounting holes 3 are used for the passing of fixing parts, so as to fix the base 2 on the ground or other positions.

[0029] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0030] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A disaster emergency assessment method integrating intelligent analysis of surveying and mapping data, characterized by: The method comprises the following steps: Step 1: Collect optical and radar images of the ground surface in different weather conditions; Step 2: pre-processing the remote sensing mapping data obtained in step 1 as well as the terrain data, building image data and environmental parameter data; For image data, perform cropping and normalization operations; For sensor data, perform data cleaning and standardization operations; Extract effective features from the preprocessed data, use the labeled sample data to train the machine learning model, use the cross-validation method to evaluate the performance of the model, optimize and adjust the model according to the evaluation results, and calculate the building damage index and road traffic obstruction coefficient and the prediction of the diffusion range of dangerous gases based on the output results of the model and the integrated auxiliary information and comprehensive assessment of the disaster situation; Step 3: The remote sensing data pre-processed in step 2 is transmitted to the disaster emergency intelligent assessment system for further processing. The damage degree of the building structures in the disaster-stricken area is identified and classified according to the existing labeled data. The terrain changes and the damage degree of the building structures are classified into two or more categories. The remote sensing mapping data is fused, and then feature extraction is performed to determine the degree of damage. The time series data is processed, and the analysis is performed according to the wind speed time series environmental parameters to predict the diffusion trend of the hazardous gas. Step 4: assign different weights to the remote sensing mapping data and meteorological monitoring data processed in step 3, and then perform weighted average fusion and dynamic data fusion to integrate the remote sensing mapping data and meteorological monitoring data in real time to obtain terrain and building information, thereby improving the real-time and accuracy of disaster assessment; Step 5: Store, manage, analyze and visualize the geospatial data processed in step 4. In disaster assessment, overlay the information on topographic changes and building structure damage in the disaster-stricken area with the population density and infrastructure layers in GIS to intuitively display the distribution and impact range of the disaster.

2. A disaster emergency assessment device integrating surveying and mapping data intelligent analysis, comprising the disaster emergency assessment method integrating surveying and mapping data intelligent analysis as claimed in claim 1, the device comprising a disaster surveying and mapping device and a disaster emergency intelligent assessment system, characterized in that: The disaster mapping device comprises a lifting column (1), an unmanned aerial vehicle (9), a synthetic aperture radar sensor (13), a millimeter wave radar (14), an optical camera (15), a meteorological sensor (7) and a gas sensor (8); a bottom plate (11) is provided at the lower end of the unmanned aerial vehicle (9), and the bottom plate (11) is connected to the unmanned aerial vehicle (9); the millimeter wave radar (14) and the optical camera (15) are installed on the left side of the lower end of the bottom plate (11); and the synthetic aperture radar sensor (13) is installed on the right side of the lower end of the bottom plate (11); The disaster emergency intelligent assessment system comprises a machine learning module (16), a data fusion module (21), a geographic information system (24), a disaster emergency assessment integrated system (25) and a disaster data processing module (31).

3. The disaster emergency assessment device integrating surveying and mapping data intelligent analysis according to claim 2 is characterized by: The disaster data processing module (31) comprises a data acquisition module (26), a data preprocessing module (27), a feature extraction module (28), a model training and evaluation module (29) and a disaster evaluation module (30).

4. The disaster emergency assessment device integrating surveying and mapping data intelligent analysis according to claim 3 is characterized by: The machine learning module (16) includes a decision tree algorithm module (17), a support vector machine (18), a convolutional neural network (19) and a recurrent neural network (20).

5. The disaster emergency assessment device integrating surveying and mapping data intelligent analysis according to claim 4 is characterized by: The data fusion module (21) comprises a weighted balancing module (22) and a Kalman filtering module (23).

6. The disaster emergency assessment device integrating surveying and mapping data intelligent analysis according to claim 5 is characterized by: A driving motor (4) is provided at the upper end of the lifting column (1), and the driving motor (4) is connected to the lifting column (1); a transparent top box (5) is installed at the output end of the driving motor (4), and the transparent top box (5) is connected to the output end of the driving motor (4).

7. The disaster emergency assessment device integrating surveying and mapping data intelligent analysis according to claim 6 is characterized by: A wireless charging receiving device (12) is provided at the lower end of the bottom plate (11), and the wireless charging receiving device (12) is connected to the bottom plate (11).

8. The disaster emergency assessment device integrating surveying and mapping data intelligent analysis according to claim 7 is characterized by: A wireless charger (10) is arranged inside the transparent top box (5) below the wireless charging receiving device (12), and the wireless charger (10) is connected to the transparent top box (5).

9. The disaster emergency assessment device integrating surveying and mapping data intelligent analysis according to claim 8 is characterized by: A fixing ring (6) is arranged outside the lifting column (1), and the fixing ring (6) is connected to the lifting column (1), and the meteorological sensor (7) and the gas sensor (8) are respectively arranged on both sides of the outside of the fixing ring (6).

10. The disaster emergency assessment device integrating surveying and mapping data intelligent analysis according to claim 9 is characterized in that: A base (2) is provided at the lower end of the lifting column (1), and the base (2) is connected to the lifting column (1); four mounting holes (3) evenly distributed along a ring are provided inside the base (2), and the mounting holes (3) and the base (2) are integrally formed.

Images